Applied optics最新文献

Traditional spectacle lens design methodologies have been hindered by their high complexity and low efficiency, primarily due to their reliance on Coddington equations or classical optimization algorithms. We propose an efficient spectacle lens design method based on differentiable ray tracing (DRT), where the partial derivatives of the merit function with respect to the lens surface parameters are computed through automatic differentiation. A -12D lens design demonstrates that the proposed method outperforms traditional optimization approaches, including simulated annealing (SA), particle swarm optimization (PSO), and genetic algorithm (GA), in terms of optimization efficiency, edge thickness, and mean power error, while achieving lower distortion compared to SA and PSO. Additionally, we explored the impact of vertex distance variation on the design results, as well as the spatial distribution of optimized lenses in relation to distortion and mean power error.

Binocular structured light technology is widely employed in high-precision 3D reconstruction due to its reliability and accuracy. However, it has inherent limitations, including partial point cloud loss resulting from limited perspectives, occlusions, and uneven illumination. To address this issue, this study proposes a monocular-binocular fusion completion framework. The approach integrates a multi-phase point cloud registration process, as well as adaptive threshold filtering and multi-step point cloud post-processing. The proposed method effectively enhances the completeness of binocular point clouds without requiring additional hardware. The feasibility of the framework is validated through theoretical analysis and empirical testing.

In satellite-to-satellite optical communications, accurate beacon tracking necessitates two optical terminals. However, during tracking, star background light noise in the receiver's charge-coupled device field of view interferes with beacon light identification, impairing system performance and communication stability, and potentially disrupting the laser communication link. The fast Fourier transform (FFT) algorithm is typically used to differentiate the beacon from starlight based on frequency characteristics. Nonetheless, when the motion trajectories of the beacon and background lights are similar and their light spots overlap, FFT has difficulty distinguishing them, causing incorrect identification of the beacon light, and then disrupting the tracking process. We introduce the fractional Fourier transform segmentation approach, which represents signals in a fractional Fourier field by rotating any angle counterclockwise around the origin in the time-frequency plane, which isolates the beacon from the background light. The method enables effective separation of the beacon light from complex scenarios, so the tracking process will not be interrupted when background light interference occurs at the terminal. This method has great value for improving the tracking performance of laser communication on the beacon light.

Smart windows play an important role in regulating solar radiation and reducing building energy consumption. Thermochromic smart windows based on VO₂ nanoparticles offer great potential for large-scale applications. However, non-uniform distributions of particle size and volume fraction along the film thickness direction are often present in VO₂ films, which not only affect optical modulation but also induce uneven local phase transitions, further impacting thermal regulation and response efficiency. In this study, a multiscale coupled model was developed using the finite-difference time-domain (FDTD) method and the finite element method (FEM) to investigate how these non-uniformities influence the optical and thermal behavior of VO₂ nanoparticle films. When the particle size gradient increased from 30-40-50 to 30-50-100 nm, τ_lum dropped by 8.29% and Δτ_sol by 7.43%, with similar trends observed across different film thicknesses. In 15 µm films, the peak temperature of the film with a uniform particle size distribution is 2.31°C higher than that of the non-uniform film, which promotes more complete and more synchronous phase transition. The influence of non-uniform volume fraction on optical performance was limited, but its effect on local temperature response was pronounced. In 5 µm films, a downward-f_v distribution increased the peak temperature by 0.75°C. When the particle size variation is within 10 nm, a uniform model remains effective; however, with larger gradients or thicker films, fine modeling is required to ensure predictive accuracy. Optimizing particle size and volume fraction distributions can improve the optical-thermal response of VO₂ films and lower the required transition temperature.

The coupled nonlinear Schrödinger equation (CNLSE) governs signal propagation in polarization division multiplexed (PDM) optical fiber systems, yet poses significant numerical challenges. This paper introduces physics-informed neural networks (PINNs) as a novel framework for deterministic modeling of PDM transmission. Through validation across single-pulse evolution, communication sequences, and full PDM systems, PINNs demonstrate deterministic accuracy (RMSE=0.0044∼0.0129 and spectralerrors<4%) while overcoming traditional limitation. They eliminate the split-step Fourier method (SSFM)'s step-size dependencies and data-driven methods' statistical uncertainties. By preserving physical determinism through embedded PDE constraints, PINNs establish a new paradigm, to our knowledge, for reliable fiber-optic system modeling.

Metasurfaces have shown great potential in optical imaging and sensing due to their ability to manipulate light fields at the subwavelength scale, especially in spectral detection. In this study, we propose a computational miniature spectrometer operating in the mid-infrared range by integrating a dielectric metasurface with a gold nanopillar array. A genetic algorithm is introduced for the first time, to our knowledge, to select the optimal combination of structures from 726 candidates to construct a filter array. By combining four conventional reconstruction algorithms with a fine-tuning optimization method, high-precision spectral reconstruction is achieved across the 3-5 µm range, with a resolution of 20 nm. This work provides a practical and scalable solution for the miniaturization and integration of mid-infrared spectrometers, offering significant potential for both research and application.

This research introduces a fingerprint-based artificial neural network approach for visible light positioning systems. The study evaluates four different light emitting diode (LED) configurations-square, rectangular, triangular, and circular-within a 5m×5m×3m indoor environment to determine which arrangement delivers the highest positioning accuracy. The analysis employs a receiver moving in a circular path within the receiver plane for the estimation of positioning accuracy across the entire trajectory. Comprehensive simulation results, including the cumulative distribution function, frequency distribution of positioning errors, and error magnitudes at different receiver locations, demonstrate that the mean positioning errors (in centimeters) are 15.2893, 12.4548, 52.5016, and 9.8749 for square, rectangular, triangular, and circular configurations, respectively. The findings indicate that the circular arrangement yields superior performance with minimal positioning error. This configuration creates consistent signal strength gradients across all directions, eliminating potential "dead zones" and maximizing line-of-sight connections between LEDs and the receiver regardless of position. The superior performance of the circular configuration underscores the significant impact of geometric arrangement on positioning accuracy, even when utilizing identical numbers of LEDs and signal processing techniques.

To address the challenge of dynamically controlling the operating bands of terahertz wave polarization conversion, we propose a tunable terahertz metamaterial device capable of switching its asymmetric transmission between single-band and dual-band modes. The device consists of a top layer of gold and vanadium dioxide, separated from the bottom layer by a polyimide spacer layer. The bottom layer is a mirror-symmetric and 90°-rotated configuration of the top structure. Leveraging vanadium dioxide's phase transition property from an insulating to a metallic state upon heating enables the tunability of incident terahertz waves. In the metallic state, the structure exhibits pronounced single-band asymmetric transmission, with a transmission coefficient T_yx reaching 0.85 at 1.54 THz, while T_xy remains at 0.06. In the insulating state, dual-band asymmetric transmission is observed, with peak T_yx values of 0.786 at 1.45 THz and 0.783 at 2.0 THz, and corresponding T_xy values of 0.06 and 0.07, respectively. Analysis of the structural surface currents reveals that the structure excites asymmetric dipole current resonance, which enables cross-coupling between the incident electric field and induced magnetic field, resulting in asymmetric transmission. The introduction of the coupled-mode theory abstracts the metamaterial as a coupled dual-resonator system, thereby providing further insights into the physical mechanism of dual-band polarization conversion. This tunable asymmetric transmission device presents a promising approach for expanding the applications of filters and tunable optoelectronic devices.

In automotive applications, lidar systems are essential both for driver assistance systems and autonomous driving, yet adverse weather conditions like fog can lead to a considerably smaller obstacle detection range. Polarized illumination in combination with polarization-aware detection has been considered as a possibility to counteract. Therefore, we investigated time-dependent light propagation in an infinite homogeneous scattering medium, with an emphasis on polarization. Analytically, we derived solutions of the vectorial radiative transfer equation for the Green's functions of the single and double scattered radiance originating from a unidirectional point source. Moreover, we implemented an electric-field Monte Carlo simulation of a bistatic scanning lidar system and found very good agreement with the analytic results.

We demonstrate the viability of pumping Yb³⁺,Tm³⁺ codoped crystals by commercial 976 nm Yb-fiber lasers for achieving low-threshold, efficient and power-scalable operation at 1.5 and 2.3 µm. A Yb,Tm:LiYF₄ laser employing this pumping scheme generated a continuous-wave output power up to 0.70 W at ∼2307nm (the ³H₄→³H₅ transition) with a slope efficiency of 26.7%, laser threshold down to 46 mW, and linear polarization (π). It also delivered 0.60 W at 1497 nm (the ³H₄→³F₄ transition) with a slope efficiency of 26.1% and a laser threshold of 78 mW. Polarization switching at 1.5 µm between π and σ states was observed and explained by gain anisotropy. Various approaches for pumping 2.3 µm Tm lasers are discussed and compared.